Figure 1
From: PFClust: an optimised implementation of a parameter-free clustering algorithm
Execution times. Comparison of the execution times between the original (black, top row) and new (grey, bottom row) implementations, averaged over the seven datasets from[9]. The different steps of the algorithm (Randomization, Clustering and Total Execution time) are shown from left to right. The combined process of randomization and clustering has to be run four times (or occasionally more[9]), the totals given here include these repetitions.